Synthesis, crystal structures, and molecular hyperpolarizabilities of a new Schiff base ligand, and its copper(II), nickel(II), and cobalt(II) metal complexes

Article abstract of DOI:10.1016/j.ica.2004.03.004

A new ligand (HL) obtained from the Schiff base condensation of 4-(diethylamino)salicylaldehyde with 4-nitroaniline is reported, with its nickel(II), copper(II), and cobalt(II) complexes. The crystal structures are reported for the four derivatives. While, Ni^IIL₂ and Cu^IIL₂ are centrosymmetric molecules, Co ^IIL₂ exhibits a pseudo-tetrahedral molecular structure. The quadratic hyperpolarizabilities (β) of HL and Co^IIL ₂, measured by electric field induced second harmonic (EFISH) technique, are equal to 66 and 110×10^-30 cm⁵esu ^-1, respectively. Beside a geometric effect (pseudo-T_d symmetry), the coordination of the metal center provides an intrinsic enhancement of the NLO response. In addition, an enhancement of the thermal stability of about 60° is found upon metal complexation.

Full text of DOI:10.1016/j.ica.2004.03.004

Inorganica Chimica Acta 357 (2004) 3825–3835
www.elsevier.com/locate/ica
Synthesis, crystal structures, and molecular hyperpolarizabilities of
a new Schiff base ligand, and its copper(II), nickel(II), and
cobalt(II) metal complexes
a,
Pascal G. Lacroix , Frederic Averseng , Isabelle Malfant , Keitaro Nakatani
a
a
b
*
ꢀ ꢀ
a
Laboratoire de Chimie de Coordination du CNRS, 205 route de Narbonne, 31077 Toulouse, France
PPSM, Ecole Normale Supꢀerieure de Cachan, URA 1906, Avenue du Pdt Wilson, 94235 Cachan, France
b
Received 12 February 2004; accepted 6 March 2004
Available online 8 April 2004
Dedicated to Professor Tobin J. Marks for his outstanding contributions to all aspects of inorganic chemistry
Abstract
A new ligand (HL) obtained from the Schiff base condensation of 4-(diethylamino)salicylaldehyde with 4-nitroaniline is reported,
with its nickel(II), copper(II), and cobalt(II) complexes. The crystal structures are reported for the four derivatives. While, NiIIL2 and
CuIIL2 are centrosymmetric molecules, CoIIL2 exhibits a pseudo-tetrahedral molecular structure. The quadratic hyperpolarizabilities (b)
of HL and CoIIL2, measured by electric field induced second harmonic (EFISH) technique, are equal to 66 and 110 ꢀ 10ꢁ30 cm5 esuꢁ1
,
respectively. Beside a geometric effect (pseudo-Td symmetry), the coordination of the metal center provides an intrinsic enhancement of
the NLO response. In addition, an enhancement of the thermal stability of about 60° is found upon metal complexation.
Ó 2004 Elsevier B.V. All rights reserved.
Keywords: Schiff bases complexes; Nonlinear optics; EFISH; Semi-empirical calculations; X-ray diffraction
1. Introduction
gating metal–organic structures for NLO purpose. The
initial approach was mainly based on a systematic
screening of the NLO response of inorganic [21] and
organometallic [22] species, within the framework of the
emerging INDO/SOS formalism [23,24]. Some synthetic
work conducted by Tobin’s group came also to imple-
ment the initial idea that a metal center can lead to
enlarged NLO response [25].
Later on, Tobin et al. investigated the promising class
of bis(salicylaldiminato)metal Schiff base complexes
[26–29]. In these systems, the metal atom is located at
the center of the chromophore, making a better use of
the d-hybridation scheme of the metal. It has been ob-
served that varying the nature of the metal could lead to
various electronic properties, with enhancement of the
NLO response in some cases, the geometry of the rigid
bis(salicylaldiminato)-based complex being roughly un-
affected. On the occasion of the special issue of Inorga-
nica Chimica Acta dedicated to the career of Professor
Marks, it seemed interesting to us to bring about the
Molecular materials with quadratic nonlinear optical
(NLO) properties are currently attracting considerable in-
terest [1–8]. The quadratic hyperpolarizability (b) of a
chromophore is associated with charge transfers between
donor and acceptor substituents linked through p-conju-
gated pathways. Most NLO chromophores have been
found in organic molecules, such as substituted stilbene
[9,10], thiophene [4,11], or polyene families [12] but the in-
troduction of a metal center as a donor or acceptor subunit
has also led to various novel organometallic [13–15] and
inorganic [16–20] molecules with large NLO responses.
In the late 1980s, Professor Tobin Marks was one of
the first researchers who became interested in investi-
* Corresponding author. Tel.: +33-5-61-33-31-88; fax: +33-5-61-55-
30-03.
E-mail address: pascal@lcc-toulouse.fr (P.G. Lacroix).
0020-1693/$ - see front matter Ó 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2004.03.004

3826
P.G. Lacroix et al. / Inorganica Chimica Acta 357 (2004) 3825–3835
heating: 10 °C/min). The decomposition temperature
N
(Td) was assigned as the intercept of the leading edge of
the decomposition endotherm with the baseline of the
DTA scan [32].
O2N
DANS
2.2. Synthesis
N
HL: In absolute ethanol (40 mL) were successfully
added 276 mg (2 ꢀ 10ꢁ3 mol) of 4-nitroaniline, 386 mg
(2 ꢀ 10ꢁ3 mol) of 4-diethylaminosalicylaldehyde, and one
drop of sulfuric acid. The mixture was refluxed for two
days. After concentration to 1/3 of the initial volume and
cooling down, a red precipitate was filtered, washed with
the minimum amount of cold absolute ethanol, and dried
under vacuum. Yield: 501 mg, (80%). Anal. Calc. for
C16H19N2O2: C, 65.16; H, 6.11; N, 13.41. Found: C,
64.90; H, 6.22; N, 13.33%. 1H NMR(CDCl3): d ¼ 1:216
(t, 7.1 Hz, 6H), 3.417 (q, 7.1 Hz, 4H), 6.221 (d, 1.8 Hz,
1H), 6.277 (dd, 2.0 and 8.9 Hz, 1H), 7.180 (d, 8.8 Hz, 1H),
7.295 (d, 8.9 Hz, 2H), 8.241 (d, 8.9 Hz, 1H), 8.408 (s, 1H),
13.160 (s, 1H). Single crystals suitable for X-ray studies
were obtained by recrystallization in ethanol.
NiL2: To a solution of 250 mg (8 ꢀ 10ꢁ4 mol) of HL
in 20 mL of hot ethanol was added, a solution of 114 mg
(4 ꢀ 10ꢁ4 mol) of Ni(AcO)2 ꢂ 4H2O dissolved in 20 mL of
hot ethanol, and the resulting solution was refluxed for
two days. After concentration to one half of the initial
volume and cooling down, an orange precipitate was
filtered, washed with ethanol, and dried under vacuum.
Yield: 205 mg (75%). Anal. Calc. for C34H36N6NiO6: C,
59.76; H, 5.31; N, 12.30. Found: C, 59.95; H, 5.02; N,
12.28%. Single crystals suitable for X-ray studies were
obtained by recrystallization in ethanol.
O2N
N
HO
HL
Scheme 1.
field of NLO bis(salicylaldiminato)metal complexes with
an additional investigation.
Contrary to the previous reports in which the metal
atom was located at the center of rigid square planar
tetradentate ligands, the present contribution explores
electronic effects obtained upon metal complexation, in
alternatives molecular geometries. This can be achieved
by changing the nature of the metal and by using a non-
rigid Schiff base ligand. 40-Dimethylamino-1-nitrostil-
bene (DANS, Scheme 1), which has widely been used as
the typical benchmark NLO chromophore along the
1980s offers a useful model skeleton for this investiga-
tion. The parent (diethylaminosalicylaldiminato)nitro-
benzene (HL, Scheme1) derivative, in which a nitrogen
atom has been incorporated in the p-conjugated bridge
to ensure the coordination of a metal, has been preferred
for the present investigation. Our study focuses on the
synthesis and crystal structures of HL and its nickel(II),
copper(II), and cobalt(II) metal complexes. The elec-
tronic properties of the different species are compared in
relation to their molecular structures, and the NLO re-
sponses are investigated by the electric field induced
second harmonic (EFISH) technique [30,31] and within
the framework of the INDO/SOS (ZINDO) formalism
[23,24].
CuL2: Following the same procedure, the use of
80 mg (4 ꢀ 10ꢁ4 mol) of Cu(AcO)2 ꢂ H2O affords
179 mg (65 % yield) of an orange precipitate. Anal.
Calc. for C34H36CuN6O6: C, 59.34; H, 5.21; N, 12.21.
Found: C, 59.55; H, 5.40; N, 12.15%. Single crystals
suitable for X-ray studies were obtained by recrystal-
lization in ethanol.
2. Experimental
CoL2: Similarly, the use of 100 mg (4 ꢀ 10ꢁ4 mol) of
Co(AcO)2 ꢂ 4H2O affords 191 mg (70% yield) of a red
precipitate. Anal. Calc. for C34H36CoN6O6: C, 59.74; H,
5.31; N, 12.29. Found: C, 59.67; H, 5.17; N, 11.97%.
Single crystals suitable for X-ray studies were obtained
as follows: in one arm of a H-shaped tube, a solution of
8 ꢀ 10ꢁ4 mol of HL in ethanol was slowly diffused into a
solution containing 4 ꢀ 10ꢁ4 mol of Co(AcO)2 ꢂ 4H2O in
the other arm of the tube. Crystals were obtained over a
period of one week.
2.1. Materials and equipment
4-Nitroaniline, Ni(AcO)2 ꢂ 4H2O, Cu(AcO)2 ꢂ H2O,
and Co(AcO)2 ꢂ 4H2O were purchased from Aldrich and
used as received, as were DANS (Kodak) and 4-(diethyl-
amino)-salicylaldehyde (Lancaster). Solvents (SDS or
Calo Erba) for the spectroscopic studies were used
without further purification. The 1H NMR spectra were
recorded on a Bruker AM 250 spectrometer. Elemental
analyses were performed by the Service de Microanal-
yses du LCC-CNRS, Toulouse. Thermal measurements
were performed by TG/DTA analysis on a Setaram-
TGDTA92 thermoanalyser. The experiments were con-
ducted under nitrogen on 5 mg of sample (rate of
2.3. X-ray data collection and structure determination
The data were collected on a Stoe Imaging Plate
Diffraction System (IPDS) equipped with an Oxford
cryosystems cooler device, with a tube power of 1.5 kW

P.G. Lacroix et al. / Inorganica Chimica Acta 357 (2004) 3825–3835
3827
for HL and NiIIL2, and 1.8 kW for CuIIL2 and CoIIL2.
The crystal to detector distance was 80 mm for the three
complexes and 70 mm for the ligand. For HL, NiIIL2
and CuIIL2, the structures were solved by direct meth-
ods (SHELXS-86) [33] and refined by least-square pro-
cedures. Crystallographic data are summarized in
Table 1. The weighting scheme used in the last refine-
factors) have been deposited with the Cambridge Crys-
tallographic Data Centre.
2.4. Spectroscopy
The UV–visible spectra were recorded on a Hewlett–
Packard 8452 A spectrophotometer. The oscillator
strengths were extracted from the spectra through the
2
2
ment cycles was w ¼ w0½1 ꢁ ðDF =6rðFoÞ Þꢃ , where
P
w0 ¼ 1= n ArTrðxÞ with coefficients Ar for the Cheby-
relation [38]:
Z
where the integration extends over the entire absorption
1
shev polynomial ArTrðxÞ, where x was Fc=FcðmaxÞ hav-
ing the following values: 1.695, )1.490, 1.523, )0.609
and 0.262 for HL; 1.346, 0.546 and 0.798 for NiIIL2;
5.698, )7.715, 4.870 and )2.342 for CuIIL2. The calcu-
lations were carried out with the CRYSTALS package
programs [34] running on a PC. The structure of CoIIL2
was solved by SHELXS-97 [35] and refined using the
least-square method on F 2. The drawings of the mo-
lecular structures were obtained with the help of
CAMERON [36]. The atomic scattering factors were
taken from International Tables for X-ray Crystallog-
raphy [37]. Crystallographic data (excluding structure
f ¼ 4:315 ꢀ 10ꢁ9 edm;
ð7Þ
band and m is the wavenumber (cmꢁ1).
2.5. NLO measurements
The quadratic molecular hyperpolarizabilities (b)
were measured by the EFISH technique [30,31]. The
light source was a nanosecond (10 ns) Nd-YAG pulsed
(10 Hz) laser (B.M. Industries) operating at k ¼ 1:064
Table 1
Crystal data, data collection and refinement for HL, NiIIL, CuIIL, and CoIIL
HL
NiIIL2
CuIIL2
CoIIL2
Crystal data
Chemical formula
Molecular weight
Crystal size (mm)
Crystal system
Space group
C17H19N3O3
313.36
0.7 ꢀ 0.4 ꢀ 0.1
monoclinic
P21=c
16.272(2)
7.900(1)
13.288(1)
90
C17H18N3Ni0:5O3
341.70
C17H18Cu0:5N3O3
344.12
C17H18Co0:5N3O3
341.81
0.2 ꢀ 0.15 ꢀ 0.1
monoclinic
C2=c
25.763(6)
9.931(4)
12.657(3)
90
0.8 ꢀ 0.55 ꢀ 0.1
0.2 ꢀ 0.1 ꢀ 0.05
triclinic
ꢁ
triclinic
ꢁ
P1
P1
ꢂ
a (A)
7.8106(12)
9.6823(15)
10.7656(15)
84.63(2)
81.65(2)
86.04(2)
800.7
7.2215(12)
8.3765(12)
13.5007(18)
77.88(2)
88.84(2)
78.94(2)
783.49
2
ꢂ
b (A)
ꢂ
c (A)
a (°)
b (°)
c (°)
114.04(1)
90
96.62(3)
90
3
ꢂ
V (A )
1559.99
4
3216.6
8
1428
Z
2
F ð000Þ
664.15
1.334
0.871
358.35
1.417
6.597
359.33
1.458
7.510
q(calcd.) (Mg/m3)
1.412
5.890
293
l(Mo Ka) (cmꢁ1
T (K)
)
180
293
160
Data collection
ꢂ
Radiation (Mo Ka) (A)
Scan mode
Scan range
0.71073
/
0 < / < 200
3:3 < 2h < 52:1
14 594
0.71073
/
0 < / < 250:5
3:8 < 2h < 48:2
6459
0.71073
/
0 < / < 250:5
5:1 < 2h < 48:3
6433
0.71073
/
0 < / < 250:6
4:40 < 2h < 46:5
13 077
2h range
Reflections collected
Reflections unique (Rm)
Reflections observed
2969 (0.058)
1392[I > 3rðIÞ]
2382 (0.02)
1997[I > 3rðIÞ]
2364 (0.11)
1131[I > 3rðIÞ]
2484 (0.13)
2230[I > 2rðIÞ]
Refinement
Refinement on
F
F
F
F 2
R
0.0324
0.0361
213
0.048
0.057
215
0.0616
0.0609
215
0.0622
0.1494
215
wR
Number of variables
ꢁ3
ꢂ
ꢂ
Dqmax (e Aꢁ3
)
0.112
)0.124
1.02
0.606
)0.306
1.07
0.599
)0.479
1.13
0.232
)0.223
0.90
Dqmin (e A
)
Goodness-of-fit

3828
P.G. Lacroix et al. / Inorganica Chimica Acta 357 (2004) 3825–3835
lm. The outcoming Stokes-shifted radiation at 1.907 lm
generated by Raman effect in a hydrogen cell (1 m long,
50 bar) was used as the fundamental beam for second
harmonic generation (SHG). Chloroform was used as a
solvent for any EFISH measurement. The centrosym-
metry of the solution was broken by dipolar orientation
of the chromophores with a high voltage pulse (5 kV)
synchronized with the laser pulse. The dipole moments
were measured independently by a classic method based
on the Guggenheim theory [39]. Further details of the
experimental methodology and data analysis are re-
ported elsewhere [40].
procedure previously used in the case of weakly reacting
amines, such as 40-nitroaniline [45,46]. The completeness
of the reaction can be easily monitored by H NMR
1
spectroscopy to verify the disappearance of the CHO
signal located at 9.47 ppm in the starting (diethyla-
mino)salicylaldehyde and the appearance of the imine
signal at 8.41 ppm. Metal complexes were obtained by
simple mixing of HL with the appropriate metal(II)
acetate in ethanol. Contrary to NiIIL2 and CuIIL2,
CoIIL2 was found to be very weakly soluble in most
common solvents, and no crystals suitable for X-ray
experiments were obtained by recrystallization. On the
other hand, large single crystals were easily obtained by
diffusion. The thermal stability of the chromophores
rises from 260 (HL) to 320 °C (CoIIL2) after metal
complexation. This observation may deserve some in-
terest, as a good thermal stability has become an im-
portant prerequisite for various practical applications.
2.6. Calculation of NLO response
The all-valence INDO/S (intermediate neglect of
differential overlap) method [41,42], in connection with
the sum-over-state (SOS) formalism [43], was employed.
Details on the computationally efficient INDO-SOS-
based method for describing second-order molecular
optical nonlinearities have been reported elsewhere
[23,24]. The calculation of electronic transitions and
molecular hyperpolarizabilities was performed using the
commercially available MSI software package
INSIGHT II (4.0.0) [44]. In the present approach, the
closed-shell restricted Hartree-Fock (RHF) formalism
was employed. The monoexcited configuration interac-
tion (MECI) approximation was employed to describe
the excited states. The 100 energy transitions between
the 10 highest occupied molecular orbitals and the 10
lowest unoccupied ones were chosen to undergo CI
mixing. Metrical parameters used for the calculations
were taken from the present crystal structure of HL. The
structure of DANS was a model obtained from the HL
backbone.
3.2. Description of the structures
Figs. 1–4 illustrate the atomic numbering scheme
employed for molecules HL, NiIIL2, CuIIL2, and
CoIIL2, respectively. The X-ray structure analysis of HL
reveals P21=c space group, with one HL entity in the
asymmetric unit. Except for the ethyl substituents of the
amine, the molecule is nearly planar, with an angle of
10°77 between the two phenyl groups. This value is in-
dicative of significant conjugation of the whole p-system
and hence large intramolecular charge transfer
upon electron excitation from the amine to the nitro-
substituent.
II
II
ꢁ
Ni L2 and Cu L2 crystallize in the triclinic P1 space
group. The asymmetric unit contains half a metal atom
and one Lꢁ entity, with the metal atom located at the
inversion center, in both cases. Therefore, NiIIL2 and
CuIIL2 are centrosymmetric molecules, which cannot
exhibit any quadratic NLO effect (b ¼ 0). For symmetry
reason, the geometry is strictly planar around the metal
centers in both molecules. Centrosymmetry is a natural
trend in nickel(II) complexes. By contrast, the geometry
of bis-bidentate Schiff base copper(II) complexes can be
the result of steric and electronic effects which result in
various intermediate geometries between square planar
3. Results and discussion
3.1. Synthesis and characterization
The HL ligand has been obtained by Schiff base
condensation in the presence of a drop of sulfuric acid, a
Fig. 1. Atom labeling scheme and thermal vibration ellipsoids for HL.

P.G. Lacroix et al. / Inorganica Chimica Acta 357 (2004) 3825–3835
3829
Fig. 2. Atom labeling scheme and thermal vibration ellipsoids for NiIIL2. H atoms are omitted for clarity.
Fig. 3. Atom labeling scheme and thermal vibration ellipsoids for CuIIL2. H atoms are omitted for clarity.
and tetrahedral [47–49]. However, no distortion is ob-
served in the coordination sphere of the present CuIIL2,
a situation that may tentatively be related to either a
modest steric hindrance in the nitrophenyl moieties or a
very negative reduction potential in the complex, ac-
cording to the analysis provided in the previous inves-
tigations [47–49].
The coordination spheres of NiIIL2 and CuIIL2 (bond
lengths and angles) are compared in Table 2, and found
rather similar. A slight shortening of the metal–ligand
bond lengths is observed on passing from the copper(II)
to the nickel(II) derivative. These structural features,
which favor steric hindrance between the two Lꢁ frag-
ments, can account for the distortion of the ligands in
both complexes. Indeed, the angles between the phenyl
groups are equal to 72.69° and 52.31°, for NiIIL2 and
CuIIL2, respectively. These observations suggest a re-
duced conjugation, versus that of the free HL ligand.
CoIIL2 crystallizes in the monoclinic C2=c space
group. The asymmetric unit contains half a metal atom
and one Lꢁ entity, with the metal atom located on the
twofold axis. By contrast with the previous NiIIL2 and
CuIIL2 complexes, CoIIL2 reveals a pseudo-tetrahedral
coordination sphere, in which the two Lꢁ fragments
refer to each other by the twofold axis. The angle be-
tween the two O(1)–Co(1)–N(1) planes of the molecule
is equal to 99.34°. The present geometry strongly re-
duces the steric hindrance and therefore the conforma-
tion of the Lꢁ entities is close to that of the free ligand.
In particular, the observation of an angle equal to 12.42°

3830
P.G. Lacroix et al. / Inorganica Chimica Acta 357 (2004) 3825–3835
Fig. 4. Atom labeling scheme and thermal vibration ellipsoids for CoIIL2. H atoms are omitted for clarity.
Table 2
II
II
II
ꢂ
Selected bond lengths (A in A) and angles (B in °) in the coordination sphere of Ni L2, Cu L2, and Co L2
(A)
Ni(1)–O(1)
Ni(1)–N(1)
Cu(1)–O(1)
Cu(1)–N(1)
Co(1)–O(1)
Co(1)–N(1)
1.815(2)
1.907(3)
1.859(5)
2.008(6)
1.859(5)
1.983(6)
(B)
O(1)–Ni(1)–O(1)#1
O(1)–Ni(1)–N(1)
O(1)–Ni(1)–N(1)
N(1)–Ni(1)–N(1)#1
O(1)–Cu(1)–O(1)#1
O(1)–Cu(1)–N(1)
O(1)–Cu(1)–N(1)
N(1)–Cu(1)–N(1)#1
180.0
O(1)–Co(1)–O(1)#2
O(1)–Co(1)–N(1)
116.9(4)
96.5(3)
93.1(1)
86.9(1)
180.0
O(1)#2–Co(1)–N(1)
N(1)–Co(1)–N(1)#2
110.1(2)
128.2(4)
180.0
90.9(2)
89.1(2)
180.0
Symmetry transformations used to generate equivalent atoms: #1 2 ꢁ x, 1 ꢁ y, 2 ꢁ z; #2 2 ꢁ x, y, 3=2 ꢁ z.
between the two phenyl groups favors the conjugation
over the whole extent of each ligand, allowing a better
use of p-electron system for NLO purpose.
both bands exhibit a large positive solvatochromism (red
shift in solvent of higher polarity). This behavior is usually
associated with changes in dipole moments between the
ground (g) and the excited states (e) upon excitation
(Dl > 0 in the case of positive solvatochromism), and is
indicative of large quadratic hyperpolarizabilities, ac-
cording to the well known two level description of the
NLO response, through the relation [51,52]:
3.3. Optical spectroscopy
The absorption spectra of HL and DANS in dichlo-
romethane are compared in Fig. 5. Both exhibit an intense
band at 434 nm (e ¼ 34900 Mꢁ1 cmꢁ1) and 439 nm
(e ¼ 26200 Mꢁ1 cmꢁ1), for HL and DANS, respectively.
The absorption maxima recorded in non-protic solvents
of different polarities are gathered in Fig. 6 versus the
Reichardt solvent parameter [50]. It can be observed that
4
2
X
3e ꢁhf Dl
ðDEÞ
ꢁꢀ
2
ꢀ
ꢁ
b ¼
ꢀ
;
3
2
2
2
2mðDEÞ
ðDEÞ ꢁ ð2ꢁhxÞ
ðDEÞ ꢁ ðꢁhxÞ
i
ð1Þ

P.G. Lacroix et al. / Inorganica Chimica Acta 357 (2004) 3825–3835
3831
4x104
8x104
7x104
6x104
5x104
4x104
3x104
2x104
1x104
0
4x104
3x104
2x104
1x104
0
(4)
(1)
3x104
(2)
(3)
2x104
1x104
0
300
350
400
450
500
550
600
300
350
400
450
500
550
600
wavelength (nm)
wavelength (nm)
Fig. 7. Electronic spectra in dichloromethane for HL (1) compared
with those of CuIIL2 (2), NiIIL2 (3), and CoIIL2 (4). The extinction
coefficients for the ligand are indicated on the left and for the com-
plexes on the right.
Fig. 5. Electronic spectra for HL in dichloromethane. DANS (dotted
line) is given as a reference.
455
450
445
440
435
430
425
420
The aforementioned observations can be compared
with the INDO-derived spectroscopic properties gath-
ered in Table 3. The INDO data are blue shifted by
about 50 nm and the oscillator strengths are overesti-
mated versus the experiment. However, the same trend
for higher energy and intensity and reduced solvato-
chromic shift on passing from DANS to HL is evidenced
both experimentally and theoretically. Therefore, we can
assume that the calculated spectra roughly fit the ex-
perimental data, in order to relate the NLO response
and the INDO-based electronic properties.
0.2
0.3
0.4
0.1
N
ET
The absorption spectra of CuIIL2, NiIIL2, and CoIIL2
are compared with that of HL in Fig. 7. The copper and
nickel derivatives exhibit reduced oscillator strengths
(per Lꢁ fragment) and blue shifts versus HL with ab-
sorption maxima equal to 431 nm (e ¼ 53100
Fig. 6. Solvatochromism versus the Reichardt empirical solvent pa-
rameter for HL. The solvatochromism of DANS (white diamond,
dotted line) is given as a reference. The slopes are equal to 60, and 73
for HL and DANS, respectively.
M
ꢁ1 cmꢁ1) and 394 nm (e ¼ 40500 Mꢁ1 cmꢁ1), respec-
in which DE is the energy of the g ! eðiÞ transition, f its
oscillator strength, and hx the energy of the incident
laser beam. In most push–pull organic chromophores,
the summation in Eq. (1) can be restricted to a single
and intense low-lying transition. The examination of
Fig. 6 may suggest that the Dl parameter for HL is
lower than that of DANS. However, it has to be re-
minded that the estimation of Dl by means of solvato-
chromism is somewhat ambiguous and not fully reliable
in some cases [53,54].
tively. These modifications can be related to the mo-
lecular structures of both complexes. As discussed in the
previous section, the planar geometry of the ligands is
hampered by steric hindrance in NiIIL2. The resulting
lack of conjugation is consistent with the changes ob-
served in the intensity and energy of the charge transfer
transition. In CuIIL2, the ligands are slightly more pla-
nar, in relation to longer metal–ligand bond lengths.
Therefore, the same qualitative modifications are evi-
denced in the electronic spectra on passing from HL to
CuIIL2, however to a less extent. In striking contrast
with the previous complexes, CoIIL2 exhibits a pseudo-
tetrahedric coordination around the metal center. As a
main consequence, no steric hindrance takes place in the
ligand, the absorption spectrum of which is therefore
closely related to that of HL. Nevertheless, the presence
of the metal leads to a slight red shift and increase of the
oscillator strength per ligand unit, with kmax ¼ 446 nm
and e ¼ 76150 Mꢁ1 cmꢁ1 (f ¼ 1:55). Both effects sug-
gest a possible b enhancement on passing from HL to
Table 3
Comparison of experimental (in dichloromethane) and INDO spec-
troscopic data (kmax in nm, e in Mꢁ1 cmꢁ1 and oscillator strength f ) for
HL and DANS
Experimental
kmax
434 0.70 60
INDO
f
Solvatochromisma
kmax
f
Dl
HL
378
397
1.35 11.6
1.16 18.4
DANS 439 0.59 73
a Slopes in Fig. 6.

3832
P.G. Lacroix et al. / Inorganica Chimica Acta 357 (2004) 3825–3835
Table 4
Molecular hyperpolarizabilities for HL and CoIIL2 compared with
DANS
the sum-over-state (SOS) perturbation theory, b is re-
lated to all excited states of a molecule [43] and can be
partitioned into two contributions, so-called b2level and
b3level terms [24]. Analysis of term contributions reveals
that in the present case, b2level (Eq. (1)) dominates the
nonlinearity, as is usually the case [24]. It can be related
to a single HOMO and LUMO-based transition (1 ! 4),
responsible for 85% and 84% of b2level for HL and
DANS, respectively (Table 5). Therefore, we can make
the assumption that understanding the changes occur-
ring for these transitions on passing from DANS to HL
can provide qualitative understanding of the origin of
the b differences for both chromophores. The orbitals
are compared in Fig. 8. The calculation reveals that the
HOMO–LUMO energy gap is enlarged and the charge
transfer reduced on passing from DANS to HL, which is
consistent with a blue shift, a reduced Dl factor, and
hence a reduced hyperpolarizability. Nevertheless, and
contrary to the intuition, the hydroxy substituent seems
not to be directly involved in the charge-transfer process
in HL.
A direct comparison of the NLO response of HL
and CoIIL2 raises the important issue of b calculations
for open-shell molecules. To the best of our knowledge,
the only ones reported with ZINDO are due to Marks
and coworkers [27,28]. However, they were restricted to
chromophores with spin states equal to 1/2. In CoIIL2,
the metal center possesses the d7 configuration, which
in tetrahedric environment corresponds to a spin state
S ¼ 3=2. In this case, the b calculation cannot be per-
formed using our ZINDO release. Nevertheless, a the-
oretical hyperpolarizability of CoIIL2 can be evaluated
from the tensorial sum of the individual hyperpolariz-
ability coefficients of the ligands. In this model, no
enhancement is expected as compared to the response
of the individual ligands. Assuming a one-dimensional
character for the NLO tensor of the subunit, b has only
one large coefficient along the charge-transfer axis (Ox),
namely bligand ¼ bxxx. In the C2 symmetry, this will lead
to a complex having the following b tensor component
[56]:
INDO data
EFISH data
l (D)
b1:097 lm
b1:097 lm
DANSa
HL
CoIIL2
6.6
6.4
7.1
73
66
70
47
b
110
–
a Ref. [10].
b No data available for open shell chromophores.
CoIIL2, according to Eq. (1). This will be discussed in
the next section.
3.4. NLO properties
NiIIL2 and CuIIL2 are centrosymmetric molecules
and, therefore, possess strictly vanishing molecular hy-
perpolarizabilities. By contrast, the b values of HL and
CoIIL2 measured by electric field second harmonic
(EFISH) technique are compared with that of DANS in
Table 4. Before discussing the effect of metal complex-
ation in the NLO response, it is interesting to compare
the properties of HL and DANS [10].
Nicoud and Twieg have pointed out that the b values
of stilbene analogs change significantly with the incor-
poration of nitrogen atoms in the p-conjugated bridge.
In particular, they have reported that benzylidene
nitroanilines exhibit reduced oscillator strengths and
blue shifts versus their nitrostilbene analogs, which re-
sults in b values divided by 3–4 [55]. Surprisingly, these
effects do not take place in the present HL ligand,
although the data clearly reveal a slight b reduction,
with values equal to 73 and 66 ꢀ 10ꢁ30 cm5 esuꢁ1, for
DANS and HL, respectively. At first, the presence of an
hydroxy substituent with donating capability in HL
might account for this effect.
In order to verify this possibility, the INDO calcu-
lated b values are provided in Table 4. Indeed, the cal-
culated data reveal that the b value of HL is only slightly
reduced versus that of DANS. Within the framework of
Table 5
Energies (kmax in nm), oscillator strengths (f ), dipole moment changes (Dl in D), contribution in the b2level, and composition of the transitions for HL
and DANS
Transition
kmax
f
Dl
State (%)a
Compositionb of CI expansion
HL
1 ! 4
1 ! 19
378
208
1.35
0.53
11.6
7.6
85
3
0:782v60!61 ꢁ 0:472v60!62
ꢁ0:577v54!61 þ 0:415v60!63
DANS
1 ! 4
1 ! 8
397
265
1.16
0.21
ꢂ
18.4
24.3
ꢃ
84
5
0:859v57!58
ꢁ0:748v56!58 ꢁ 0:412v57!58
bg!eðiÞ
i bg!eðiÞ
a Contribution of the ith transition to b2level state % ¼
.
P
b Orbital 60 is the HOMO and orbital 61 the LUMO for HL. Orbital 57 is the HOMO and orbital 58 the LUMO for DANS.

P.G. Lacroix et al. / Inorganica Chimica Acta 357 (2004) 3825–3835
3833
With the assumption that the NLO response of CoIIL2 is
the sum of the contributions of the ligand subunits in C2
symmetry (Fig. 9), the charge-transfer axis is the same
for HL and for the organic subunits in CoIIL2. There-
fore, the ZINDO calculated b for HL and the crystal
data for CoIIL2 lead to h ¼ 43°. This provides the fol-
lowing relationship:
92%
86%
LUMO
(E = -1.48 eV)
LUMO
(E = -1.51 eV)
bðcomplexÞ ¼ 1:46 ꢀ bðligandÞ
:
ð6Þ
This expression has to be compared to the experimental
data (Table 3), which indicates bðcomplexÞ ¼ 1:67bðligandÞ
.
Within the approximation of this model, a b enhance-
ment of 14% can be inferred upon metal complexation.
It is interesting to compare this b enhancement with
those previously reported in other Schiff base complexes.
Several reports of metal salen-based NLO chromo-
phores have pointed out that large b enhancements
(more than twice the hyperpolarizabilities of the free
ligands in some cases) can be obtained in planar ge-
ometries [29,45,59]. In these previous investigations, the
enhancements were not related to geometric effects but
to electronic effects, only. By contrast, the present b
enhancement (bðcomplexÞ ¼ 1:67bðligandÞ) arises from a
combination of geometric and, to a lesser extent, elec-
tronic effects. These observations seems to indicate that
tetrahedral geometry in bis(salicylaldiminato)metal
Schiff base complexes can provide an alternative route
towards efficient enhancements of the NLO response at
the microscopic level.
70%
70%
HOMO
HOMO
(E = -7.13 eV)
(E = -7.57 eV)
Fig. 8. Frontier orbitals with their related energies for HL (left) and
DANS (right). The electron densitities (%) are indicated on the dom-
inant fragments.
bZYY ðcomplexÞ ¼ 2bxxx cos h sin2 h;
bZZZðcomplexÞ ¼ 2bxxx cos3 h
ð2Þ
ð3Þ
all other components of the tensor are weak (h is defined
as the angle between the charge-transfer axis (Ox) of the
subunit and the twofold axis (OZ) of the metal complex,
according to Fig. 9). Within the framework of the
EFISH method, the measured b value is the projection
of the hyperpolarizability along the dipole moment (C2
axis in Fig. 9). Therefore, b ¼ bZ. With a laser operating
at 1.907 lm, one can assume that no absorption takes
place in the materials. Therefore, the Kleinman condi-
tions can be applied [57,58]:
4. Conclusion
A Schiff base ligand has been presented with several
of its metal(II) complexes. Depending on the nature of
the metal center, the geometry of the resulting molecule
can be tuned from square planar to pseudo-tetrahedral.
While the nickel(II) and copper(II) derivatives are cen-
trosymmetric, and therefore exhibit vanishing hyperpo-
larizabilities, the experimental data indicate that
bZ ¼ bZXX þ bZYY þ bZZZ
which leads to
;
ð4Þ
b
is equal to 1.67b
for CoIIL2. Beside the
ðcomplexÞ
ðligandÞ
bZ ¼ 2bxxx cos3 h þ 2bxxx cos h sin2 h:
ð5Þ
effect of the coordination on the overall charge-transfer
geometry, a b enhancement of 14% is evidenced, related
to the electronic effect induced by the metal center, while
the thermal stability rises from 260 to 320 °C.
C2
Several bis(salicylaldiminato) metal complexes with
quadratic NLO properties have been appeared in the
literature [18,19]. It has previously been reported that
the metal can lead to b enhancement by use of the d
orbitals [45] and b modulation by magnetic transition
occurring at the metal center [60]. The present study
suggests that cobalt(II) metal complexes, which exhibit a
trend towards tetrahedral geometry, can lead to ex-
tended chromophores with noncentrosymmetric molec-
ular geometry and high thermal stability. These
structural features could also provide a route towards
βxxx
θ
Z
O
Y
Fig. 9. Model geometry for the CoIIL2 chromophore.

3834
P.G. Lacroix et al. / Inorganica Chimica Acta 357 (2004) 3825–3835
ꢃ
[26] S. Di Bella, I. Fragala, I. Ledoux, M.A. Diaz-Garcia, P.G.
Lacroix, T.J. Marks, Chem. Mater. 6 (1994) 881.
ꢃ
[27] S. Di Bella, I. Fragala, I. Ledoux, T.J. Marks, J. Am. Chem. Soc.
extended NLO architectures, or chromophores with
octupolar geometries [61].
117 (1995) 9481.
[28] S. Di Bella, I. Fragala, T.J. Marks, M.A. Ratner, J. Am. Chem.
ꢃ
Soc. 118 (1996) 12747.
5. Supplementary material
ꢃ
[29] S. Di Bella, I. Fragala, I. Ledoux, M.A. Diaz-Garcia, T.J. Marks,
J. Am. Chem. Soc. 119 (1997) 9550.
[30] B.F. Levine, C.G. Betha, J. Chem. Phys. 63 (1975) 2666.
[31] B.F. Levine, C.G. Betha, J. Chem. Phys. 65 (1976) 1989.
[32] C.R. Moylan, R.J. Twieg, V.Y. Lee, S.A. Swanson, K. Betterton,
R.D. Miller, J. Am. Chem. Soc. 115 (1993) 12599.
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic
Data Centre, CCDC Nos. 231301 (HL), 231302
(NiIIL2), 231303 (CuIIL2), and 231304 (CoIIL2). Copies
of this information may be obtained free of charge from:
The Director, CCDC, 12 Union Road, Cambridge, CB2
1E2, UK (fax: +44-1223-336-033; e-mail: deposit@ccdc.
cam.ac.uk, or www:http://www.ccdc.cam.ac.uk).
[33] G.M. Sheldrick, SHELXS-86, Program for Crystal Structure
€
€
Solution, University of Gottingen, Gottingen, Germany, 1986.
[34] D.J. Watkin, C.K. Prout, J.R. Carruthers, P.W. Betteridge,
CRYSTALS
, Issue 10, Chemical Crystallography Laboratory,
University of Oxford, Oxford, England, 1996.
[35] G.M. Sheldrick, SHELX97 – Programs for Crystal Structure
€
€
Analysis (Release 97-2), Institut fur Anorganische Chemie der
€
€
Universitat, Tammanstrasse 4, D-3400 Gottingen, Germany,
1998.
References
[36] D.J. Watkin, C.K. Prout, L.J. Pearce, CAMERON, Chemical
Crystallography Laboratory, University of Oxford, Oxford, En-
gland, 1996.
[1] P.N. Prasad, D.J. Williams, Introduction to Nonlinear Optical
Effects in Molecules and Polymers, Wiley, New York, 1991.
[2] J. Zyss, Molecular Nonlinear Optics: Materials, Physics, and
Devices, Academic Press, New York, 1994.
[37] D.T. Cromer, J.T. Waber, in: International Tables for X-Ray
Crystallography, vol. 4, Kynoch Press, Birmingham, 1974.
ꢀ
[38] M. Orchin, H.H. Jaffe, Symmetry Orbitals, and Spectra, Wiley,
New York, 1971, p. 204.
[3] H.S. Nalwa, S. Miyata, Nonlinear Optics of Organic Molecules
and Polymers, CRC Press, New York, 1997.
[39] E.A. Guggenheim, Trans. Faraday Soc. 45 (1949) 714.
[40] I. Maltey, J.A. Delaire, K. Nakatani, P. Wang, X. Shi, S. Wu,
Adv. Mater. Opt. Electronics 6 (1996) 233.
[4] L.R. Dalton, A.W. Harper, R. Ghosn, W.H. Steier, M. Ziari, H.
Fetterman, Y. Shi, R.V. Mustacich, A.K.-Y. Jen, K.J. Shea,
Chem. Mater. 7 (1995) 1060.
[41] M.C. Zerner, G. Loew, R. Kirchner, U. Mueller-Westerhoff, J.
Am. Chem. Soc. 102 (1980) 589.
[5] R.G. Benning, J. Mater. Chem. 5 (1995) 365.
[6] D.M. Burland (Ed.), Optical Nonlinearities in Chemistry, Chem.
Rev. 94(1) (1994).
[42] W.P. Anderson, D. Edwards, M.C. Zerner, Inorg. Chem. 25
(1986) 2728.
[43] J.F. Ward, Rev. Mod. Phys. 37 (1965) 1.
[7] T. Verbiest, S. Houbrechts, M. Kauranen, K. Clays, A. Persoons,
J. Mater. Chem. 7 (1997) 2175.
[44] M.J. Frisch, G.W. Trucks, H.B. Schlegel, P.M.W. Gill, B.G.
Johnson, M.A. Robb, J.R. Cheeseman, T. Keith, G.A. Petersson,
J.A. Montgomery, K. Raghavachari, M.A. Al- Laham, V.G.
Zakrzewski, J.V. Ortiz, J.B. Foresman, J. Cioslowski, B.B.
Stefanov, A. Nanayakkara, M. Challacombe, C.Y. Peng, P.Y.
Ayala, W. Chen, M.W. Wong, J.L. Andres, E.S. Replogle, R.
Gomperts, R.L. Martin, D.J. Fox, J.S. Binkley, D.J. Defrees, J.
Baker, J.P. Stewart, M. Head-Gordon, C. Gonzalez, J.A. Pople,
Gaussian 94, Revision E.2, Gaussian, Inc., Pittsburgh, PA, 1995.
[45] P.G. Lacroix, S. Di Bella, I. Ledoux, Chem. Mater. 8 (1996) 541.
[8] J.J. Wolff, R. Wortmann, Adv. Phys. Org. Chem. 32 (1999) 121.
[9] L.T. Cheng, W. Tam, S.H. Stevenson, G.R. Meredith, G. Rikken,
S.R. Marder, J. Phys. Chem. 95 (1991) 10631.
[10] L.T. Cheng, W. Tam, S.R. Marder, A.E. Stiegman, G. Rikken,
C.W. Spangler, J. Phys. Chem. 95 (1991) 10643.
[11] V. Pushkana Rao, A.K.Y. Jen, J. Chandrasekhar, I.N. Nam-
boothiri, A. Rathna, J. Am. Chem. Soc. 118 (1996) 12443.
[12] M. Blanchard-Desce, V. Alain, P.V. Bedworth, S.R. Marder, A.
Fort, C. Runser, M. Barzoukas, S. Lebus, R. Wortmann, Chem.
Eur. J. 3 (1997) 1091.
€
[46] D. Wohrle, P. Buttner, Polym. Bull. (Berlin) 13 (1985) 57.
[13] S.R. Marder, in: D.W. Bruce, D. O’Hare (Eds.), Inorganic
Materials, Wiley, New York, 1997, p. 121.
[47] J.M. Fernandez-G, S. Hernandez-Ortega, R. Cetina-Rosado, N.
Macias-Ruvalcaba, M. Aguilar-Martinez, Polyhedron 17 (1998)
2425.
[14] N.J. Long, Angew. Chem., Int. Ed. Engl. 34 (1995) 21.
[15] I.R. Whittall, A.M. McDonagh, M.G. Humphrey, M. Samoc,
Adv. Organomet. Chem. 42 (1998) 291.
[48] J.M. Fernandez-G, M.R. Patino-Maya, R.A. Toscano, L. Ve-
lasco, M. Otero-Lopez, M. Aguilar-Martinez, Polyhedron 16
(1997) 4371.
[16] S.M. LeCours, H.W. Guan, S.G. DiMagno, C.H. Wang, M.J.
Therien, J. Am. Chem. Soc. 118 (1996) 1497.
[49] M. Aguilar-Martinez, R. Saloma-Aguilar, N. Marcias Ruvalcaba,
R. Cetina-Rosado, A. Navarrete-Vasquez, V. Gomez-Vidales, A.
Zentella-Dehesa, R.A. Toscano, S. Hernandez-Ortega, J.M.
Fernandez-G, J. Chem. Soc., Dalton Trans. (2001) 2346.
[17] H. Lebozec, T. Renouard, Eur. J. Inorg. Chem. (2000) 229.
[18] S. Di Bella, Chem. Soc. Rev. 30 (2001) 355.
[19] P.G. Lacroix, Eur. J. Inorg. Chem. (2001) 339.
[20] F. Cariati, U. Caruso, R. Centore, A. De Maria, M. Fusco, B.
Panunzi, A. Roviello, A. Tuzi, Inorg. Chim. Acta 357 (2004) 548.
[21] D.R. Kanis, M.A. Ratner, T.J. Marks, Chem. Mater. 3 (1991) 19.
[22] D.R. Kanis, M.A. Ratner, T.J. Marks, J. Am. Chem. Soc. 114
(1992) 10338.
€
[50] C. Reichardt, E. Harbusch-Gornet, Liebigs Ann. Chem. 5 (1983)
721.
[51] J.L. Oudar, J. Chem. Phys. 67 (1977) 446.
[52] J.L. Oudar, J. Chemla, J. Chem. Phys. 66 (1977) 2664.
ꢀ
[53] V. Alain, S. Redoglia, M. Blanchard-Desce, S. Lebus, K.
[23] D.R. Kanis, T.J. Marks, M.A. Ratner, Int. J. Quant. Chem. 43
(1992) 61.
Lukaszuk, R. Wortmann, U. Gubler, C. Bosshard, P. Gu€nter,
Chem. Phys. 245 (1999) 51.
[24] D.R. Kanis, M.A. Ratner, T.J. Marks, Chem. Rev. 94 (1994) 195.
[25] D.R. Kanis, P.G. Lacroix, M.A. Ratner, T.J. Marks, J. Am.
Chem. Soc. 116 (1994) 10089.
[54] M. Blanchard-Desce, V. Alain, L. Midrier, R. Wortmann, S.
Lebus, C. Glania, P. Kramer, A. Fort, J. Muller, M. Barzoukas, J.
Photochem. Photobiol. A 195 (1997) 115.

P.G. Lacroix et al. / Inorganica Chimica Acta 357 (2004) 3825–3835
3835
[55] J.F. Nicoud, R.J. Twieg, in: D.S. Chemla, J. Zyss (Eds.),
Nonlinear Optical Properties of Organic Molecules and
Crystals, vol. 1, Academic Press, New York, 1987,
p. 277.
[58] D.A. Kleinman, Phys. Rev. 128 (1962) 1761.
[59] F. Averseng, P.G. Lacroix, I. Malfant, G. Lenoble, P. Cassoux, K.
Nakatani, I. Maltey- Fanton, J.A. Delaire, A. Aukauloo, Chem.
Mater. 11 (1999) 995.
[56] T. Thami, P. Bassoul, M.A. Petit, J. Simon, A. Fort, M.
Barzoukas, A. Villaeys, J. Am. Chem. Soc. 114 (1992) 915.
[57] D.A. Kleinman, Phys. Rev. 126 (1962) 1977.
[60] F. Averseng, C. Lepetit, P.G. Lacroix, J.P. Tuchagues, Chem.
Mater. 12 (2000) 2225.
[61] J. Zyss, I. Ledoux, Chem. Rev. 94 (1994) 77.